This Week in Allergy & Immunology — Aug 30, 2026

Welcome to This Week in Allergy & Immunology. This week we're covering 10 notable papers spanning anaphylaxis risk and peanut allergy, the widening debate over asthma biologics and biomarkers, and new approaches to immunotherapy delivery and allergen monitoring. Let's dive in.

We start with a finding that may change how you risk-stratify a very common patient. In the Journal of Allergy and Clinical Immunology: In Practice, Sohng and colleagues asked whether lipid-lowering drugs worsen anaphylaxis, based on the mechanistic premise that platelet-activating factor acetylhydrolase circulates bound to low-density lipoprotein, so lowering LDL may blunt the body's ability to inactivate platelet-activating factor [1]. Across 778 patients from Canada, the United States and Germany evaluated between 2010 and 2023 for multisystem reactions to insect stings, 111 patients, about one in seven, were on a lipid-lowering medication. After adjusting for age, sex, beta-blockers, ACE inhibitors, respiratory disease and mast cell disorders, statin or other lipid-lowering therapy roughly doubled the odds of severe rather than mild-to-moderate anaphylaxis, and the effect held in the subgroup of 518 patients with venom allergy confirmed by skin or serum testing. A sensitivity analysis substituting baseline tryptase in the roughly 200 patients with measurements pointed the same direction but was not statistically significant, so this is an association from retrospective chart review rather than proof of causation. Still, the practical implication is concrete. When you see a patient with Hymenoptera venom allergy on a statin, treat that as a potential severity modifier alongside the beta-blocker and mast cell questions you already ask, and let it lower your threshold for venom immunotherapy, modified build-up protocols, extended treatment duration, and insistence on carrying epinephrine.

Staying with anaphylaxis but moving to peanut, two papers address opposite ends of the pipeline. In Annals of Allergy, Asthma and Immunology, Mustafa and colleagues took on the logistical problem that has limited infant peanut oral immunotherapy, namely that every dose escalation traditionally happens in the office [3]. Fifty-three children with a median age of 13 months completed a protocol in which updosing occurred through a mix of in-office, video and telephone encounters. Fourteen children, about a quarter, had a total of 24 treatment-related reactions; over half of the reactions resolved on their own, most of the rest settled with antihistamines, and a single reaction required epinephrine, which notably occurred during an in-office visit. Only three reactions happened during updosing at all, translating to roughly a 0.4 percent risk of reaction per dose, and the higher odds of reaction with updosing versus home maintenance did not reach statistical significance. Nearly three quarters of caregivers preferred the hybrid model, citing convenience and less travel. This is a small single-protocol cohort, so it is not license to abandon in-office supervision entirely, but it supports a hybrid remote updosing model as a way to expand access to infant peanut oral immunotherapy in practices with limited slots. Then in Allergy, Cosey and colleagues mapped where patient IgE actually binds on Ara h 1 [10]. Probing peanut peptide microarrays with sera from 75 allergic patients, the frequently recognised regions included the unstructured region in over nine in ten patients and the amino-terminal alpha-hairpinin domain in nearly nine in ten, comparable to or higher than the classic carboxy-terminal vicilin region. Human monoclonal antibodies segregated cleanly between the two domains without meaningful cross-reactivity, and one anti-alpha-hairpinin antibody blocked the majority of patient IgE against that domain. Functionally, paired antibodies against alpha-hairpinin failed to trigger passive systemic anaphylaxis in mice, while a single antibody against the trimeric vicilin region did. The message for diagnostics is that current component testing may be missing a dominant epitope, and including the alpha-hairpinin domain could sharpen peanut allergy diagnosis.

Turning to asthma, the World Allergy Organization Journal published an analysis that puts hard numbers on a question every severe asthma clinic argues about, namely who should be eligible for biologics [2]. Yang and colleagues used nationally representative United Kingdom primary care and hospital records covering more than 1.3 million people with active asthma from 2004 to 2021. Just over seven percent had severe asthma, and under current United Kingdom criteria requiring severe asthma plus at least three exacerbations, only about a fifth of those with severe asthma qualified, which is 1.5 percent of all asthma patients. Applying broader global criteria roughly tripled eligibility, and applying the entry criteria used in the pivotal randomised trials, which allowed moderate asthma, increased it about six-fold to as high as nine percent of all asthma. Among those with eosinophil data, roughly four in ten met anti-interleukin-5 criteria. The budget consequences are stark: about 261 million pounds over five years under United Kingdom criteria, up to 767 million under global criteria, and as much as 1.6 billion under trial criteria. The authors also note that the narrowly eligible United Kingdom group carried more lower respiratory infection, pneumonia and bronchiectasis, a reminder that these are genuinely sick patients. This is the trade-off to bring to formulary conversations: a large population of patients with severe asthma sits just outside the eligibility line, and closing that gap is clinically defensible but not cost-neutral.

Selecting patients better is the natural companion question, and in Allergology International, Fujisawa and colleagues reviewed biomarkers in paediatric asthma [7]. Their conclusion is deliberately sobering. Exhaled nitric oxide and blood eosinophil count remain the only clinically mature markers, useful for identifying type 2 inflammation and supporting decisions about anti-inflammatory or biologic therapy, with IgE sensitisation adding phenotypic context for anti-IgE selection. Induced sputum gives direct airway phenotyping but stays confined to specialist centres. Eosinophil-derived neurotoxin may reflect eosinophil activation rather than just eosinophil count, but needs assay harmonisation and paediatric reference ranges. Everything else, from periostin and microRNAs to breath, metabolomic, proteomic and microbiome signatures, remains exploratory and is not yet clinically actionable. If you are being marketed a novel paediatric asthma biomarker panel, that is the yardstick to apply.

Occupational airway disease is the third asthma thread, and two papers converge on the same clinical discipline. In JACI: In Practice, Suojalehto and colleagues followed up 618 patients with work-related asthma who had undergone specific inhalation challenge, with 274 responding a median of eight and a half years later [5]. Nearly four in five patients reported no current exposure to the suspected causal agent, and three quarters of working-aged patients were still in the workforce. About a third of patients had difficult-to-treat asthma at follow-up, and clinical remission off treatment was rare, seen in only four percent. Critically, long-term outcomes were essentially the same whether the inhalation challenge had been positive or negative, though positive patients were more likely to have changed occupation. What predicted a poor long-term course was not the challenge result but the state of the asthma at diagnosis: older age, an asthma control test score below 20, requiring GINA step four or five treatment, and already having difficult-to-treat disease, which increased the odds of a bad outcome several-fold. In other words, a negative inhalation challenge should not reassure you about prognosis. Complementing this, Current Allergy and Asthma Reports published a review by Hutson and colleagues on poultry industry workers, who are exposed to a complex mix of feather and mite allergens, feed dust, moulds, ammonia, endotoxin and particulate matter, with rising rates of occupational asthma and chronic obstructive lung disease [9]. Both papers point the same way: take the occupational history early, and follow the poorly controlled patients closely regardless of challenge results.

Finally, three papers on making allergy care more practical. Allergology International reviewed transgenic rice-based peptide immunotherapy for Japanese cedar pollinosis, which affects more than 30 million people in Japan [6]. Seven dominant human T-cell epitopes from Cry j 1 and Cry j 2 were engineered into a hybrid peptide expressed in rice, where the protein body acts as a natural delivery system protecting the peptide from gastric degradation and delivering it to gut-associated lymphoid tissue. In small clinical studies, eating the rice suppressed T-cell proliferation, modulated cytokines and improved medication scores and quality of life without treatment-related allergic adverse events, and a retrospective follow-up suggests benefit may persist after stopping. These are small-scale studies, so treat this as a proof of concept for epitope-based oral tolerance rather than an available therapy. On the environmental side, International Archives of Allergy and Immunology reported a colorimetric guanine-based rapid test for house dust mite allergen, validated against ELISA on 80 dust samples from 59 homes in Guangzhou [8]. Correlation with ELISA was strong, every sample above the 10 microgram per gram threshold was detected, specificity was around 79 percent, and agreement between patients and researchers performing the test was excellent. It needs only 20 milligrams of dust and gives a result in five minutes, which makes it plausible as a tool for patients to verify that avoidance measures are actually working. And in JACI: In Practice, Huang and colleagues reviewed women's health in eosinophilic esophagitis [4], noting that women present with less dysphagia and fewer strictures but more abdominal pain, nausea, reflux and chest pain, with symptom burden and quality-of-life impairment similar or worse despite milder endoscopic activity. Estrogen appears protective, which may partly explain male predominance. In pregnancy, symptoms often improve then recur after delivery, and proton pump inhibitors, swallowed budesonide and dupilumab appear safe to continue, while elimination diets warrant caution and dietitian supervision.

If you only have time for one paper this week, make it the lipid-lowering medication and venom anaphylaxis study in JACI: In Practice [1]. It adds a common, easily identified medication class to the short list of severity modifiers you screen for in every venom-allergic patient, and it costs nothing to act on.

Here are the key takeaways from this week in Allergy and Immunology. First, ask about statins and other lipid-lowering drugs in venom-allergic patients and treat their use as a marker of higher severity risk. Second, hybrid remote updosing for infant peanut oral immunotherapy looks safe in early data and is strongly preferred by caregivers. Third, current biologic eligibility rules exclude around four in five patients with severe asthma, and broadening them is clinically justifiable but expensive. Fourth, in paediatric asthma, exhaled nitric oxide, blood eosinophils and IgE sensitisation remain the only markers ready for clinical use. Fifth, a negative specific inhalation challenge in work-related asthma does not predict a better long-term course; baseline asthma control does.

That's your roundup for This Week in Allergy & Immunology. The full transcript and references are available on the episode page in your AudioScholar library. This is an AI-curated summary — for clinical decisions, always consult primary sources and current guidelines. See you next week.

If this weekly briefing is useful, follow the show in your podcast app so new episodes arrive automatically. And if it saves you time, a quick rating in your podcast app helps other physicians find the show.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.